INDEX

Explanations

specific numerical references and positions in documents

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

egend

-0.17

ui

-0.16

istra

-0.16

omba

-0.15

 Ritch

-0.15

ÙĪÙĩ

-0.14

erm

-0.14

asser

-0.14

łģ

-0.13

avr

-0.13

POSITIVE LOGITS

Ðĭ

0.15

obia

0.15

onym

0.14

adia

0.14

enie

0.14

posables

0.14

+++

0.14

arness

0.14

illez

0.14

-ground

0.14

Activations Density 0.002%

specific numerical references and positions in documents

No Comments

No Known Activations